This Content Downloaded From 137.132.123.69 On Tue, 18 Oct 2022 04:19:04 UTC
This Content Downloaded From 137.132.123.69 On Tue, 18 Oct 2022 04:19:04 UTC
Author(s): Osvaldo E. Sala, F. Stuart Chapin III, Juan J. Armesto, Eric Berlow, Janine
Bloomfield, Rodolfo Dirzo, Elisabeth Huber-Sanwald, Laura F. Huenneke, Robert B.
Jackson, Ann Kinzig, Rik Leemans, David M. Lodge, Harold A. Mooney, Martín
Oesterheld, N. LeRoy Poff, Martin T. Sykes, Brian H. Walker, Marilyn Walker and
Diana H. Wall
Source: Science , Mar. 10, 2000, New Series, Vol. 287, No. 5459 (Mar. 10, 2000), pp. 1770-
1774
Published by: American Association for the Advancement of Science
REFERENCES
Linked references are available on JSTOR for this article:
https://www.jstor.org/stable/3074591?seq=1&cid=pdf-
reference#references_tab_contents
You may need to log in to JSTOR to access the linked references.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms
American Association for the Advancement of Science is collaborating with JSTOR to digitize,
preserve and extend access to Science
year (-15), so we assumed that largest impact factor to temperate forests,diversity, minimize the probability of suc-
all biomes
would experience the same changeborealin CO2 arctic, and alpine. Biodiversitycessful establishment by invaders in undis-
forests,
in deserts
concentration. Nitrogen deposition and tropical forests may respondturbed communities (25). Conversely, we ex-
is largest
least cities
in the northern temperate zone near to nitrogen
and deposition because plant pect the greatest effect of biotic exchange in
is smallest in biomes such as the arctic and growth is strongly limited by water and phos- biomes such as Mediterranean and southern
southern temperate forests, which generally phorus, respectively (23). Grasslands, savan-temperate forests that have long been isolated
are distant from sources of pollution. Othernas, and Mediterranean systems received in-and exhibit extensive convergent evolution
biomes are intermediate, with regional varia-termediate impact factors because nitrogen(26). Other biomes are intermediate in their
tion in deposition generally associated withand other factors limit plant growth. connectedness. There is wide variation within
cities or industrial point sources. Climate is A given change in climate is expected tomost biomes in the successful establishment
expected to warm most dramatically at highhave the largest proportional effect on biodiver-of biotic introductions, depending on the
latitudes (arctic and boreal zones), to changesity in those biomes characteristic of extremeoriginal diversity and isolation from similar
least in the tropics, and to show intermedi-climates, although biodiversity in all biomeshabitats. For example, islands typically have
ate changes in other biomes (9) (Table 1). likely will be sensitive to climate. Small chang- low diversity and are more prone to biotic
Changes in precipitation are uncertain and are es in temperature or precipitation in arctic, al- invasions (27).
difficult to generalize at the biome level. The pine, desert, and boreal forest will result in large When averaged across biomes, land-use
pattern of biotic exchange reflects the pattern changes in species composition and biodiver-change is the driver that is expected to have
of human activity. Remote areas with little sity. Similarly, we assume that biomes wherethe largest global impact on biodiversity by
human intervention receive fewer exotic spe- climate less strongly limits the activity of or-the year 2100 (Fig. 1), mostly because of
cies than areas that are in the middle of trade ganisms will experience changes in the distri-its devastating effects on habitat availability
routes or that host intense human activity (16). bution of organisms, but the overall effect onand consequent species extinctions. Climate
The second step of our exercise was to proportional change in diversity may be less change will be the second most important
evaluate, for each biome, the impact that a pronounced than in extreme environments. driver of biodiversity change, mostly as a
unit change in each driver has on biodiversity Biotic introductions (that is, successful result of the expected warming at high lati-
independently of the expected magnitude of establishment of exotic species) vary accord- tudes. Changes in atmospheric CO2, biotic
change in the driver (Table 2). Land-use ing to environmental conditions and biogeo-exchange, and nitrogen deposition also will
change is the most severe driver of changes in graphic considerations. Invasions have oc- have substantial effects on future biodiver-
biodiversity (17). For example, conversion of curred least frequently in arctic and alpinesity, with the relative importance being re-
temperate grasslands into croplands or tropi- ecosystems, because of their severe environ- gionally variable. Variability among biomes
cal forests into grasslands results in local ment (24) and the broad longitudinal distri-of the impact of the different drivers is max-
extinction of most plant species and the as- bution of much of the high-latitude flora andimal for land use, reflecting the broad range
sociated animals whose habitat is largely de- fauna. In the tropics, we also expect a smallof expected changes in this driver and the
termined by plant species composition. Be- proportional change in the diversity of intactlarge sensitivity of all biomes to land-use
low-ground organisms are also affected most ecosystems because of the high initial diver- change. In contrast, atmospheric CO2 showed
severely by land-use change (18). We as- sity and because abiotic and biotic factors the smallest variability of the three drivers
sumed no differences among biomes in the characteristic of this biome, including its high because CO2 is well mixed in the atmosphere
response to a unit change in land use and we
assigned land use the maximum impact factor Table 1. Expected changes for the year 2100 in the five major drivers of biodiversity change (land use,
because of the consistently large effect of land- atmospheric composition CO2, nitrogen deposition, climate, and biotic exchange) for the principal
use change on biodiversity. The increase in terrestrial biomes of the Earth (arctic tundra, alpine tundra, boreal forest, grasslands, savannas, Medi-
terranean ecosystems, deserts, northern temperate forests, southern temperate forests, and tropical
atmospheric CO2 is expected to have the largest
forests).
effect on biodiversity in those biomes where
plant growth is most limited by water availabil-
ity and where there is a mixture of C3 and C4 Arctic Alpine Boreal Med Desert N Tropic
land vanna temp temp
species because of known species differences in
the effect of CO2 on water-use efficiency (19, Land use 1.0 1.0 2.0 3.0 3.0 3.0 2.0 1.0 4.0 5.0
20). For example, changes in atmospheric CO2 Climate 5.0 3.0 4.0 2.0 2.0 2.0 2.0 2.0 2.0 1.0
Nitrogen deposition 1.0 3.0 3.0 3.0 2.0 3.0 2.0 5.0 1.0 2.0
may change the competitive balance between
Biotic exchange 1.0 1.0 2.0 3.0 3.0 5.0 3.0 3.0 2.0 2.0
species that differ in rooting depth, photosyn-
Atmospheric CO2 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5
thetic pathway, or woodiness as well as associ-
ated below-ground organisms (21). Therefore,
we assigned the maximum impact factor of Table 2. Impact of a large change in each driver on the biodiversity
elevated CO2 to grasslands and savannas, change of the driver was defined for land use as conversion of 50%
which are water-limited biomes with a mixture as a 2.5-fold increase in elevated CO2 as projected by 2100, for nitro
of contrasting plant functional types. Based on for climate as a 4?C change or 30% change in precipitation, and for
the same reasoning, we assigned the smallest new plant or animal species by 2100. Estimates vary from low (1
global scenarios of the physical environment and knowledge from
impact factors to arctic, alpine, boreal forest,
tropical forest, and freshwater ecosystems.
Increased nitrogen deposition should have Arctic Alpine Boreal land vann Med Desert Tropic
land vanna temp temp
the largest impact on biodiversity in those
biomes that are most nitrogen-limited primar- Land use 5.0 5.0 5.0 5.0 5.0 5.0 5.0 5.0 5.0 5.0
Climate 4.0 4.0 3.5 3.0 3.0 3.0 4.0 2.0 2.0 3.0
ily by giving a competitive advantage to plant
Nitrogen deposition 3.0 3.0 3.0 2.0 2.0 2.0 1.0 3.0 3.0 1.0
species with high maximum growth rates,
Biotic exchange 1.0 1.0 1.0 2.0 2.0 3.0 2.0 1.5 3.0 1.5
which then exclude the slower growing spe-
Atmospheric CO2 1.0 1.0 1.0 3.0 3.0 2.0 2.0 1.5 1.5 1.0
cies (22). Consequently, we assigned the
temperate forest show large changes, mostly 0.6: actions among causes of biodiversity change.
due to changes in land use with relatively
0.2- In all scenarios, we project that grasslands
small effects due to other drivers. Arctic eco- and Mediterranean ecosystems will experi-
Alpine Desert
1-
Alpine 1 i Desert ence large biodiversity loss because of their
0.6- 0.6
1.2 sensitivity to all drivers of biodiversity
I change, particularly land-use change (Figs. 2
U 1.0 o 1- Boreal 1 N Temp and 3). We did not generate these scenarios
?0.6- 06
'o 0.8 Table 3. Expected changes for the year 2100 in
the major drivers of biodiversity change for lakes
0.6 1- Grassland 11 STemp and streams.
L 0.6-2 0.6
Uncertainties
This analysis highlights the sensitivity of
biodiversity change to our assumptions about
interactions among causes of biodiversity
change. Which assumptions are most plausi-
ble? There is clear evidence for nonlinearities
and synergistic interactions among many of
the global change drivers. Invasions of exotic
species are promoted by human disturbance Fig. 3. Maps of three scenarios of the expected change in biodiversity for the year 2100. (A) There are
and changes in climate variability (interaction no interactions among drivers of biodiversity change; consequently, total change is calculated as the
sum of the effects of each driver, which in turn result from multiplying the expected change in the driver
of biotic exchange, land-use change, and cli-
for a particular biome (Table 1) times the impact of the driver, which is also a biome-specific
mate change). Elevated CO2 has the greatest
characteristic (Table 2). (B) Total biodiversity change equals the change resulting from the driver that
effect on species composition in the presence is expected to have the largest effect and is calculated as the maximum of the effects of all the drivers.
of nitrogen deposition (interaction of CO2 (C) Interactions among the drivers are synergistic; consequently, total change is calculated as the
and nitrogen deposition). Synergistic interac- product of the changes resulting from the action of each driver. Different colors represent expected
tions may decrease in importance at extreme change in biodiversity from moderate to maximum for the different biomes of the world ranked
according to total expected change. Numbers in parentheses represent total change in biodiversity
values of individual drivers of biodiversity
relative to the maximum value projected for each scenario. Biomes are Mediterranean ecosystems
change. For example, where land use has
(MED), grasslands (GRAS), savannas (SAV), boreal forest (BOR), southern temperate forest (S.TEMP),
been severe and extensive such as in forest
tropical forest (TROP), northern temperate forest (N.TEMP), arctic ecosystems (ARCT), and desert
clearing followed by seeding of an exotic(DESERT). Values for alpine, stream, and lake ecosystems are not shown.
future changes in drivers, as thoroughly ana- man, Eds. (Springer-Verlag, Berlin, 1996), pp. 153-172.
tion of the rate of change of the drivers at the
26. H. A. Mooney and E. L. Dunn, Evolution 24, 292
lyzed by Intergovernmental Panel onscale
global Climate
and development of management (1970).
Change (38). This reflects future policies
practices specifically tailored for each region 27. P. M. Vitousek, L Loope, H. Adsersen, Eds., Islands:
governing (i) the intensity and aerial extent
according to its of
biological, social, and econom- Biological Diversity and Ecosystem Function (Springer-
Verlag, Berlin, 1995).
land-use change, (ii) the protection of biodi-
ic characteristics.
28. W. Scott Overton, in Ecosystem Modeling in Theory
versity per unit of land-use change, and (iii) and Practice: An Introduction with Case Studies,
changes in atmospheric composition. Uncer-
References and Notes C. A. S. Hall and J. W. Day, Eds. (Wiley, New York,
1. S. I. Pimm, G. J. Russell, J. L. Gittelman, T. M. Brooks, 1977), pp. 49-74.
tainties in future climate and vegetation re-
Science 269, 347 (1995). 29. Committee on Inland Aquatic Ecosystems, Water and
flect the same policy uncertainties 2. We (38).
define As a
change in biodiversity at the biome level Science Technology Board, Commission on Geo-
result of the large policy-related uncertainties as the changes in number and relative abundance sciences, of Environment, and Resources, Freshwater
species that occur naturally in that biome.
in drivers, we emphasize that we have pre- Ecosystems (National Academy Press, Washington,
3. P. M. Vitousek, Ecology 75, 1861 (1994). DC, 1996).
sented scenarios rather than predictions of
4. F. S. Chapin et al., Science 277, 500 (1997). 30. S. L. Postel, G. C. Daily, P. R. Ehrlich, Science 271, 785
biodiversity change. 5. D. Tilman et al., Science 277, 1300 (1997). (1996).
We expect large regional variation in 6. G. C. Daily, Ed., Nature's Services. Societal Depen- 31. D. M. Lodge et al., Aust. J. Ecol. 23, 53 (1998).
dence on Natural Ecosystems (Island Press, Washing-
biodiversity change within each biome. For ton, DC, 1997). 32. D. W. Schindler, BioScience 48, 157 (1998).
33. A. Ricciardi and J. Rasmussen, Conserv. Biol. 13, 1220
example, diversity in islands, lakes, and some7. R. Costanza et al., Nature 387, 253 (1997).
(1999).
streams is particularly vulnerable to biotic 8. B. H. Walker and W. Steffen, Eds., Global Change and
34. J. S. Harding, E. F. Benfield, P. V. Bolstad, G. S.
Terrestrial Ecosystems (Cambridge Univ. Press, Cam-
exchange because geographic isolation has Helfman, E. B. D. Jones, Proc. Natl. Acad. Sci. U.S.A.
bridge, 1996).
led to local adaptation and, in the case of 95, 14843 (1998).
9. A. Kattenberg et al., in Climate Change: The IPCC
35. B. D. Richter, D. P. Braun, M. A. Mendelson, L. L.
islands, often a low biodiversity (27, 31). Hot Scientific Assessment, J. T. Houghton et al., Eds. (Cam-
Master, Conserv. Biol. 11, 1081 (1997).
spots of diversity such as riparian corridors bridge Univ. Press, Cambridge, 1996), pp. 285. 36. M. W. Oswood, A. M. Milner, J. G. Irons, in Global
10. D. S. Wilcove, D. Rothstein, J. Dubow, A. Phillips, E.
and coastal land margins often coincide with Losos, BioScience 48, 607 (1998). Climate Change and Freshwater Ecosystems, P. Firth
hot spots of development, leading to large 11. T. D. Sisk, A. E. Launer, K. R. Switky, P. R. Ehrlich, and S. G. Fischer, Eds. (Springer-Verlag, New York,
BioScience 44, 592 (1994). 1992), pp. 192-210.
biodiversity loss. Within-biome variation in
12. A. Haxeltine and I. C. Prentice, Global Biogeochem. 37. N. L. Poff et al., BioScience 47, 769 (1997).
climate may cause the greatest biodiversity Cycles 10, 693 (1996). 38. T. Houghton et al., Eds., Climate Change 1995: The
change near climatically determined bound-13. J. Alcamo, Image 2: Integrated Modeling of Global Science of Climate Change (Cambridge Univ. Press,
Climate Change (Kluwer Academic, Dordrecht, Neth- Cambridge, 1996).
aries of organism distribution. Other specific
erlands, 1994). 39. F. S. Chapin III, 0. E. Sala, E. Huber-Sanwald, Eds.,
local patterns of biodiversity change are less 14. R. G. Bailey, Ecoregions: The Ecosystem Geography of Future Scenarios of Global Biodiversity (Springer-Ver-
predictable than general trends at larger the Oceans and Continents (Springer-Verlag, New lag, New York, in press).
scales and may reflect interactions among York, 1998). 40. Supported by University of California, Santa Barbara,
15. I. Y. Fung, C. J. Tucker, K. C. Prentice, . Geophys. Res. National Center for Ecological Analysis and Synthe-
drivers of biodiversity change that are locally 92D, 2999 (1987). sis, InterAmerican Institute for Global Change Re-
important or a consequence of local "surpris- 16. J. A. Drake et al., Biological Invasions: A Global Per- search, National Science Foundation OCE-9634876,
es." An initial analysis of the causes of re- spective (Wiley, Chichester, UK, 1989), vol. 37. Consejo Superior de Investigaciones Cientificas y
17. 0. E. Sala, in Global Biodiversity Assessment, Section Tecnicas, University of Buenos Aires, and Fondo para
gional variation in diversity loss within each 5, H. A. Mooney, J. Lubchenco, R. Dirzo, O. E. Sala, Eds. la Investigaci6n Cientifica y Tecnolbgica de la Agen-
biome is being published separately (39). (Cambridge Univ. Press, Cambridge, 1995). cia Nacional de Promocibn Cientifica y Tecnolbgica.
Biodiversity in all biomes is sensitive to 18. J. M. Anderson, in Global Biodiversity Assessment: This exercise stems from an activity of the Global
Section 6, H. A. Mooney, J. Lubchenco, R. Dirzo, O. E. Change and Terrestrial Ecosystems (GCTE) core
global changes in environment and land use and
Sala, Eds. (Cambridge Univ. Press, Cambridge, 1995), project of the International Geosphere-Biosphere
realistic projections of biodiversity change will pp. 406-412. Programme (IGBP). We thank A. T; Austin, D. Tilman,
require an integrated effort by climatologists, 19. H. A. Mooney, B. G. Drake, R. J. Luxmoore, W. C. and W. Reid for their useful suggestions. A. E. Sala
Oechel, L. F. Pitelka, BioScience 41, 96 (1991). and J. P. Guerschman provided valuable assistance.
ecologists, social scientists, and policy makers 20. R. B. Jackson, 0. E. Sala, C. B. Field, H. A. Mooney,
to improve scenarios of future changes in the Oecologia 98, 257 (1994). 23 August 1999; accepted 7 December 1999
Enhance your AAAS membership with the All the information you need.....in one convenient location.
Science Online advantage. . -........- ..... . Visit Science Online at http://www.scienceonline.org,
* Full text Science-research papers and news articles ?* Research Alerts-sends you an e-mail alert every time call 202-326-6417, or e-mail [email protected]
with hyperlinks from citations to related abstracts in a Science research report comes out in the discipline, or by for more information.
other journals before you receive 'Science in the mail. a specific author, citation, or keyword of your choice.
* ScienceNOW-succinct, daily briefings, of the hottest e Science's Professional Network-lists hun- AAAS is also proud to announce site-wide institutional
scientific, medical, and technological news. dreds of job openings and funding sources worldwide subscriptions to Science Online. Contactyour subscription
* Science's Next Wave-career advice, topical that are quickly and easily searchable by discipline, posi-
forums, discussion groups, and expanded news written tion, organization, and region.
by today's brightest young scientists across the world. * Electronic Marketplace-provides new product
0 ^C! ^^.^^, * &.B l gB ginformation from the world's leading science manufac- AMERICAN ASSOCIATION FOR THE
SciCenic nO N L N E turers and suppliers, all at a click of your mouse. ADVANCEMENT OF SCIENCE